Fix #913416 (Updated recipe for Pagina 12)

This commit is contained in:
Kovid Goyal 2012-01-08 20:24:06 +05:30
parent 4ce59b4817
commit ef56290eca

View File

@ -1,5 +1,5 @@
__license__ = 'GPL v3' __license__ = 'GPL v3'
__copyright__ = '2008-2010, Darko Miletic <darko.miletic at gmail.com>' __copyright__ = '2008-2012, Darko Miletic <darko.miletic at gmail.com>'
''' '''
pagina12.com.ar pagina12.com.ar
''' '''
@ -66,9 +66,7 @@ class Pagina12(BasicNewsRecipe):
return url.replace('http://www.pagina12.com.ar/','http://www.pagina12.com.ar/imprimir/') return url.replace('http://www.pagina12.com.ar/','http://www.pagina12.com.ar/imprimir/')
def get_cover_url(self): def get_cover_url(self):
rawc = self.index_to_soup('http://www.pagina12.com.ar/diario/principal/diario/index.html',True) soup = self.index_to_soup('http://www.pagina12.com.ar/diario/principal/diario/index.html')
rawc2 = re.sub(r'PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN','PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN"',rawc)
soup = BeautifulSoup(rawc2,fromEncoding=self.encoding,smartQuotesTo=None)
for image in soup.findAll('img',alt=True): for image in soup.findAll('img',alt=True):
if image['alt'].startswith('Tapa de la fecha'): if image['alt'].startswith('Tapa de la fecha'):
return image['src'] return image['src']